Abstract: The disclosure relates to technology for pre-fetching data. An apparatus comprises a processor core, pre-fetch logic, and a memory hierarchy. The pre-fetch logic is configured to generate cache pre-fetch requests for a program instruction identified by a program counter. The pre-fetch logic is configured to track one or more statistics with respect to the cache pre-fetch requests. The pre-fetch logic is configured to link the one or more statistics with the program counter. The pre-fetch logic is configured to determine a degree of the cache pre-fetch requests for the program instruction based on the one or more statistics. The memory hierarchy comprises main memory and a hierarchy of caches. The memory hierarchy further comprises a memory controller configured to pre-fetch memory blocks identified in the cache pre-fetch requests from a current level in the memory hierarchy into a higher level of the memory hierarchy.

Abstract: An application server may receive an input indicating a recovery priority for recovering data from a data backup environment to a data source environment and may receive data usage statistics indicating data access metrics and user access metrics corresponding to the data in the data source environment. The application server may generate, from the recovery priority and the data usage statistics, one or more data priority classifications for the data and may build a data model indicating an order for recovery of the data based on the one or more data priority classifications. The application server may then cause display of an indication of a progress of recovering the data from the data backup environment to the data source environment.

Abstract: Profile switching for memory overclocking is described. In accordance with the described techniques, a memory is operated according to a first memory profile. During operation of the memory according to the first memory profile, a request is received to operate the memory according to a second memory profile. Responsive to the request, operation of the memory is switched to operate according to the second memory profile without rebooting. In one or more implementations, at least one of the first memory profile or the second memory profile comprises an overclocking memory profile that configures the memory to operate in an overclocking mode. In one or more implementations, the memory is trained to operate according to the overclocking memory profile prior to operating the memory according to the first memory profile.

Abstract: A memory system having memory components, a remote direct memory access (RDMA) network interface card (RNIC), and a host system, and configured to: allocate a page of virtual memory for an application; map the page of virtual memory to a page of physical memory in the memory components; instruct the RNIC to perform an RDMA operation; perform, during the RDMA operation, a data transfer between the page of physical memory in the plurality of memory components and a remote device that is connected via a computer network to the remote direct memory access network interface card; and at least for a duration of the data transfer, lock a mapping between the page of virtual memory and the page of physical memory in the memory components.

Abstract: An administrative terminal receives designation of generation target data and of a generation destination storage device and identifies data similar to the target data. The terminal calculates a first predicted time expected for transmitting the target data from a storage device holding the target data to the generation destination storage device, and a second predicted time expected to be required for a second transmission process of transmitting the similar data from an object storage service to the generation destination storage device and of transmitting difference data between the target data and the similar data from the storage device holding the target data to the generation destination storage device. If the second predicted time is shorter than the first predicted time, the administrative terminal performs the second transmission process to transmit the similar data and the difference data to the generation destination storage device to generate the generation target data therein.

Abstract: A method performed by a processing device receives a plurality of write operation requests, where each of the write operation requests specifies a respective one of the memory units, identifies one or more operating characteristic values, where each operating characteristic value reflects one or more memory access operations performed on a memory device, and determines whether the operating characteristic values satisfy one or more threshold criteria. Responsive to determining that the operating characteristic values satisfy the one or more threshold criteria, the method performs a plurality of write operations, where each of the write operations writes data to the respective one of the memory units, and performs a multiple-read scan operation subsequent to the plurality of write operations, where the multiple-read scan operation reads data from each of the memory units.

Abstract: Circuitry comprises packet reception circuitry to receive a data communication packet with a storage classification from sending circuitry, the data communication packet including at least payload data and a target address for storage of the payload data; and storage control circuitry to control writing of the payload data of a given data communication packet by one or more storage devices selected from a set of two or more candidate storage devices each addressable by the target address, the storage control circuitry being responsive to the storage classification received with the given data communication packet and to respective persistence properties associated with the set of two or more candidate storage devices.

Abstract: Methods, systems, and devices for dynamic status registers array are described. An apparatus may include one or more memory dice coupled with a data bus. The apparatus may further include a controller coupled with each of the memory dice via the data bus that is configured to transmit a first command associated with a first operation to a first memory die. The first command may assign an associated operation (e.g., the first operation) to a queue slot of a status bank that is associated with at least the first memory die. The controller may further transmit second command to the first memory die to request a status of the first operation. The controller may receive a status of the first operation via a channel (e.g., a first channel) of the data bus that is based on the assigned queue slot of the status bank.

Abstract: A data processing system includes a memory system, a processor and a cache system. The cache system includes a cache and a data encoder associated with the cache. The data encoder encodes blocks of uncompressed data having a particular data size for storing in the memory system. The processor is configured, when an array of data has a data size equal to the particular data size or is able to be combined with one or more other arrays of data already written to the cache to provide a plurality of arrays of data having a data size that is equal to the particular data size, to output the array of data from the processor to the data encoder, bypassing the cache, for encoding as or as part of a block of data having the particular data size.

Abstract: A memory system may include: a nonvolatile memory device comprising a plurality of memory regions; and a controller in communication with the nonvolatile memory device to control operations of the nonvolatile memory device and configured to: receive a first write request including a first logical address and a second logical address; determine a duplicate physical address mapped to the second logical address; and selectively map the first logical address to the duplicate physical address based on a duplicate count corresponding to the duplicate physical address.

Abstract: Systems and methods for durable storage of storage volume “snapshots” are provided. Snapshots are stored as collections of snapshot data objects. To improve the durability of snapshot storage, physical deletion of snapshot data objects may be delayed for a period of time after the snapshot data objects are marked for deletion. Lists of the stored snapshot data objects and the snapshot data objects that make up active snapshots may be periodically analyzed. If there are any snapshot data objects that are part of active snapshots and are not present in the list of stored snapshot data objects, the snapshot data objects may be recovered before they are physically deleted.

Abstract: Techniques for management of data storage in distributed storage systems are provided. A method may include receiving, by a computer system, a request to write data to a volume. The method may include identifying, by the computer system, a zone segment mapped to the volume. The zone segment may include a plurality of zones. The method may include identifying, by the computer system, a segment pointer indicating a write location in a zone of the zone segment. The method may include writing, by the computer system, the data to one or more zones of the plurality of zones of the zone segment, starting at the write location. The method may also include updating, by the computer system, the segment pointer according to a data endpoint of the data in the zone segment.

Abstract: A system includes a memory device and a processing device, operatively coupled with the memory device, to perform operations including: receiving a request to perform a read operation on a segment of the memory device; determining a program erase cycle count associated with the segment of the memory device; determining a temperature offset value for the segment of the memory device based on a write temperature and a read temperature, determining whether the temperature offset value satisfies a threshold criterion associated with the program erase cycle count of the segment; and responsive to determining that the temperature offset value satisfies the threshold criterion, performing a corrective read operation on the segment of the memory device, wherein a sense time parameter of the corrective read operation is modified according to the temperature offset value and the program erase cycle count.

Abstract: The present technology relates to an electronic device. According to the present technology, a computing system may include a storage device and a host. The storage device may include a plurality of zones. The host may receive storage area information including an optimal write size of an open zone among the plurality of zones from the storage device, determine a target size of data to be flushed to the storage device based on the optimal write size, a history size that is a size of data previously flushed to the storage device, and a buffer data of the host, and flush data having the target size among the buffer data to the storage device.

Abstract: Methods, systems, and devices for adaptive block mapping are described. In some examples, a first superblock and a second superblock may be established across one or more dice of a memory device. The superblocks may each include one or more blocks from a plurality of planes of a memory die. In some examples, the second superblock may include at least one bad block (e.g., defective block) in addition to one or more good blocks (e.g., non-defective blocks). The memory device may receive a command for writing data in a first mode and may write a first subset of the data to the first superblock in the first mode, a second subset of the data to the second superblock in the first mode, and one or more blocks associated with the second superblock in a second mode. Additionally or alternatively, the memory device may receive a second command for writing data in the second mode and may write the data to the first superblock in the first mode.

Abstract: In an example, an apparatus comprises a plurality of compute engines; and logic, at least partially including hardware logic, to detect a cache line conflict in a last-level cache (LLC) communicatively coupled to the plurality of compute engines; and implement context-based eviction policy to determine a cache way in the cache to evict in order to resolve the cache line conflict. Other embodiments are also disclosed and claimed.

Abstract: A method of a flash memory controller to be used in a storage device and coupled to a flash memory device of the storage device through a specific communication interface includes: using a set-feature signal, which carries a set-feature command, a macro execution feature address, and corresponding macro execution parameter information, as a macro execution signal and transmitting the macro execution signal to the flash memory device to make the flash memory device execute multiple set-feature operations respectively having unique information defined by the corresponding macro execution parameter information carried in the macro execution signal.

Abstract: Aspects described herein relate to a method comprising: receiving a request to write data to a persistent storage device, the request comprising data; determining an affinity of the data; writing the request to a cache line of a cache; associating the cache line with the affinity of the data; and reporting the data as having been written to the persistent storage device.

Abstract: Methods, systems, and devices for managed memory systems with multiple priority queues are described. Memory access commands may be received from a host and stored in a command queue. First and second subsets of the commands, respectively associated with first and second priorities, may be determined. The first and second subsets may be routed from the command queue to first and second queues, respectively. The first and second subsets may be processed from the first and second queues to third and fourth queues, respectively, at a storage controller, according to first and second processes that may be run concurrently according to parameters for prioritization between the first and second priorities. Data associated with the commands may be received from the host, temporarily stored in a buffer, then moved to a storage memory (for write commands) or retrieved from the storage memory, temporarily stored in the buffer, then transmitted to the host (for read commands).

Abstract: A storage device is configured to manage a plurality of nonvolatile memories with a plurality of physical streams. An operation method of the storage device includes receiving an input/output request from an external host device, determining a 0-th virtual stream identifier, extracting a 0-th representative value from a 0-th virtual stream feature, extracting a first and second representative values corresponding to first and second physical streams, calculating distance information including first and second similarities between the 0-th virtual stream and each of the first and second physical streams, based on the extracted representative values, assigning one of the plurality of physical streams to the 0-th virtual stream, based on the distance information, and performing an operation corresponding to the input/output request, at the assigned physical stream, and the extracting and the calculating are performed by using machine learning model.